Error using reshape.

[Diego Malo]

Hello everyone,

I have the following dataset:

* Example generated by -dataex-. For more info, type help dataex
clear
input int Year float(cocoa coffee tea)
1960 3.0769715  4.229965  5.419287
1961  2.452238  4.009644  5.043255
1962 2.3037343  3.809979  4.795938
1963   2.83032  3.661893  4.921299
1964  2.555764 4.5401278 4.6623535
1965 1.8284953  4.200999  5.066413
1966  2.498808 4.0256944  4.413181
1967  2.862174  3.836209 4.7431984
1968 3.4799714  3.894003  4.034715
1969 4.1133976  3.680444  3.297254
1970  2.907652 4.4372315  3.823453
1971 2.2035952  3.939136  3.692863
1972  2.418084  3.953175 3.0436306
1973  3.661782  4.003855  2.710703
1974 4.1471906 3.6465025  3.020196
1975 2.9806585 3.3346746  2.622963
1976  4.834464  7.042325   2.87251
1977  8.285694  11.04199 4.5912213
1978  6.402001  6.434844  3.024581
1979  5.552967  6.305437  2.750204
1980  3.991594  5.143401  2.785964
1981  3.184666  3.911836  2.487348
1982  2.739556 4.3495383  2.586328
1983  3.437219 4.5747886  3.103582
1984  3.972359  5.160637  4.658217
1985  3.777082  4.925164 2.6613636
1986 3.0147066  5.489634   2.48686
1987 2.6508186  3.159186  1.741554
1988 1.9779524  3.191044 1.6969632
1989 1.5584383  2.538469 1.9059527
1990 1.5316404  1.906847 1.7972807
1991 1.4591677 1.7981176  1.744895
1992 1.3184253 1.4102768 2.0120442
1993 1.2942352  1.574493  1.804751
1994 1.6678215  3.540815 1.8767424
1995  1.558678  3.320475 1.4161226
1996 1.6137103   2.49454   1.57755
1997 1.8844392  3.436564 2.3460581
1998 2.0393958   2.92208  2.310131
1999 1.4085815 2.3445048 2.2334988
2000 1.1379292  1.779329 2.5484505
2001 1.3952132 1.2926186 1.9799826
2002  2.348558 1.3331517  1.971048
2003 2.1995945 1.4006797 1.9392273
2004 1.8233513 1.5100135 1.8285294
2005 1.7538422  2.079065 1.6847206
2006 1.7707478 2.2310436 2.1716413
2007 2.0464036 2.4281344  1.745225
2008 2.5069225 2.6276815 2.1571958
2009 2.9935236  2.494976  2.610938
2010  3.133004  3.028018  2.560042
2011  2.684754 3.7764544  2.449512
2012   2.17048  2.893732 2.6138964
2013 2.2231965 2.3481514 2.1867213
2014  2.830166 3.0685036   1.89007
2015  3.202293  2.792256  3.028274
2016  3.075278  2.959683 2.5759604
2017 2.0857253  2.851354  3.052986
2018 2.2509995 2.3529108  2.530545
2019 2.3525846 2.2620804 2.2223198
2020  2.420699  2.471659 2.0505192
2021 2.2180362  2.967605 1.9303017
2022 2.1675537  3.584937   2.21719
end

I would like to reshape this dataset to obtain the just three columns; Year, Products and Price. To make it clear, for each year, I would like to have three observations, cocoa, coffee and tea, with his corresponding price.

I am trying the following:

reshape long cocoa coffee tea, i(Year) j(products)

But I am obtaining the following error:

"no xij variables found
You typed something like reshape wide a b, i(i) j(j).
reshape looked for existing variables named a# and b# but could not find any. Remember this picture:"


Anyone knows why? Thank you.

[Nick Cox]

As documented, e.g. in the help or the manual entry, reshape long expects shared stubnames, which are prefixes (or exceptionally suffixes) shared by a bunch of variable names.

You are feeding the command variable names.

You can make progress by supplying the stub that reshape long needs.

rename (cocoa-tea) (price=)

reshape long price, i(Year) j(products) string

[Diego Malo]

Thank you Nick Cox . I readed the manually and I did not understall well, I thought that putting the name of the variable I wanted to reshape work.

[Nick Cox]

That's a common guess, but it's hard to see where it comes from. Nothing in the help supports that idea. Right from the start, the explanation of reshape long is phrased in terms of stubs.

For the advice that you may need to rename, see also https://www.stata.com/support/faqs/d...-with-reshape/

reshape has undeservedly a bad reputation, judging by grumbles on social media, but consider Stata's point of view: it needs to know exactly what you want, including new variable names as well as existing names.

It is not mentioned nearly so much, but stack is here an alternative, and may be easier to grasp. But it's worth persevering with reshape.

[Diego Malo]

Nick Cox Thank you also for the explanation! Very helpful!